OneCAD: One Classifier for All image Datasets using multimodal learning
This addresses the issue of model architecture dependency on dataset classes for researchers and practitioners in computer vision, though it appears incremental as it builds on existing methods like MIM.
The paper tackles the problem of creating a classifier model architecture that is independent of the number of classes in the dataset, proposing OneCAD, a framework using Mask-Image-Modeling with multimodal learning to achieve close-to number-of-class-agnostic performance, with preliminary results shown on natural and medical image datasets.
Vision-Transformers (ViTs) and Convolutional neural networks (CNNs) are widely used Deep Neural Networks (DNNs) for classification task. These model architectures are dependent on the number of classes in the dataset it was trained on. Any change in number of classes leads to change (partial or full) in the model's architecture. This work addresses the question: Is it possible to create a number-of-class-agnostic model architecture?. This allows model's architecture to be independent of the dataset it is trained on. This work highlights the issues with the current architectures (ViTs and CNNs). Also, proposes a training and inference framework OneCAD (One Classifier for All image Datasets) to achieve close-to number-of-class-agnostic transformer model. To best of our knowledge this is the first work to use Mask-Image-Modeling (MIM) with multimodal learning for classification task to create a DNN model architecture agnostic to the number of classes. Preliminary results are shown on natural and medical image datasets. Datasets: MNIST, CIFAR10, CIFAR100 and COVIDx. Code will soon be publicly available on github.